The FairValueLabs Team
The People Behind the Numbers
FairValueLabs is run by two people, not a content farm. Every page is reviewed by a named analyst before it publishes. The methodology is ours, the mistakes are ours, and the editorial calls are ours — including the ones we get wrong.
David Chen
Lead Equity Analyst · Co-Founder
Before FairValueLabs, David spent eight years on the buy-side at a long-only equity fund, covering industrials, consumer staples, and mid-cap financials. His seat was the "ugly duckling" desk: companies the sell-side had largely given up on, where the work was reading 10-Ks until the actual story emerged. Most of those names didn't make it into the portfolio. The ones that did paid for the rest.
He left in 2024 because the work he loved — patient, primary-source, single-name fundamental research — was getting harder to do for a paying audience that wasn't already wealthy. Bloomberg terminals start at $24,000 a year. The data underneath them, however, is free. SEC EDGAR posts every 10-K, 10-Q, 8-K, proxy, and dividend declaration the moment it's filed. The gap between "free public data" and "polished retail-facing analysis" is mostly a software problem, not a knowledge problem.
That's the thesis behind FairValueLabs. David writes the editorial frame: what the Z-Score is actually telling you about a balance sheet, why a deep-discount-to-fair-value isn't always a buy, when the moat scorecard breaks. Marcus writes the code that makes 350 ticker pages possible without ghostwriting them.
Coverage focus
- Industrials and capital goods (GE-derivative names, machine-tool makers, defense primes)
- Consumer staples and dividend aristocrats (the unloved compounders)
- Mid-cap financials and insurers (where Z-Score fails and capital ratios actually matter)
- Special situations: spin-offs, restructurings, SOTP-driven names
Education & credentials
- BBA, University of Michigan, Ross School of Business
- CFA Charterholder (passed all three levels on first attempt)
- Series 7 and 63 (lapsed since leaving the buy-side)
How he reads a stock
"The Z-Score and the moat rating are the easy parts — they're math. The hard part is figuring out which of the 200 things the 10-K tells you actually matter for the next five years. Most analysts get distracted by what's loud. The good ones notice what's quiet but persistent: a customer concentration that's been creeping up, a working-capital pattern that says revenue is being pulled forward, a buyback that's funded by debt rather than free cash flow. The numbers on this site are the table stakes. The editorial frame is where the work happens."
David is the named analyst-of-record on roughly half the ticker coverage — generally the larger-cap industrial, financial, and dividend-paying names. He reviews every fair-value, moat, and dividend-safety call before it ships.
Marcus Wei
Lead Engineer · Co-Founder
Marcus is the engineering half of FairValueLabs. Ten years in production data systems — most recently building ML inference pipelines that handled millions of inputs per minute — taught him that financial data is just another data engineering problem with worse documentation. SEC EDGAR's XBRL feed is one of the most under-loved public datasets in finance. The same balance-sheet data Bloomberg sells you, but you have to write the parser yourself. So he wrote the parser.
He's been a value investor in a personal account since 2014 — long enough to have ridden the GE collapse, the energy bear of 2015, the 2020 COVID crash, and the 2022 SaaS revaluation. Long enough to know which lessons from The Intelligent Investor still hold (margin of safety, owner earnings, Mr. Market) and which need updating for an environment with negative real rates and 30-year buyback programs. He keeps a separate paper portfolio of "FairValueLabs picks" — names where the model said buy and his judgment said yes — and publishes the misses alongside the hits.
The line between "AI-generated content" and "AI-assisted analysis" matters to him personally. The first is what's flooding the internet right now; the second is what FairValueLabs is. Every model output on this site is computed deterministically from primary financial data, then editorially reviewed by a human. If you find a number that doesn't match the underlying SEC filing, email him — he wrote the bug.
Engineering scope
- SEC EDGAR ingestion: XBRL parsing, fact normalization, financial-statement reconstruction
- Valuation models: predicted-EPS DCF, ROE-anchored P/B for financials, Altman Z, moat scorecard
- i18n localization across English, Chinese, Japanese, Korean, Spanish
- The static-site generator (Eleventy + Cloudflare Pages) and IndexNow push pipeline
Education & background
- BS Computer Science, Carnegie Mellon University
- 10 years in production ML/data infrastructure (recommendation systems, search, fraud detection)
- Personal investor since 2014; primary discipline: deep-value with quality screen
How he thinks about the models
"Every model on this site is wrong somewhere. The Z-Score doesn't apply to banks. DCF breaks for early-stage names. Moat ratings can't see disruption coming. Our job isn't to pretend the models are perfect — it's to make the assumptions visible enough that you can disagree with them in a specific way. If the methodology page tells you exactly which inputs go into the fair value, you can re-run it with your own numbers in your head. That's the difference between analysis and a black box."
Marcus is the named analyst-of-record on the technology, communication services, and speculative-growth coverage — names where the engineering and quant work tends to dominate the qualitative read. He owns the methodology page and the model-audit logs.
Editorial standards
- Named analyst-of-record on every ticker page. The byline is the person who reviewed the editorial frame, not just whoever ran the script.
- Last-reviewed timestamp on every page. When you see "Last reviewed by [name] on [date]," that means the page was read top-to-bottom by that analyst on that day, not just regenerated.
- Primary sources for every number. Every fair value, Z-Score, moat rating, and dividend safety grade traces back to a specific SEC EDGAR filing. The methodology page documents which fields go into which formulas.
- No undisclosed positions. Where the analyst-of-record holds the stock, the disclosure runs at the top of the page. Where they don't, that's stated too.
- Public miss log. When our fair value or moat rating turns out to be meaningfully wrong over a 12-month window, we publish the postmortem rather than quietly editing the page. The list lives at /about/miss-log/.
- No "buy" or "sell" recommendations. The model produces a fair value and a margin of safety. The editorial frame explains what the numbers mean. The investment decision is yours.
Reach out
Spotted a calculation error? Disagree with a moat rating? Have a name you'd like covered? David and Marcus both read incoming mail at [email protected]. We don't promise individual replies, but every flagged calculation error gets investigated and either fixed or explained on the page.